Skip to content

Latest commit

 

History

History
206 lines (170 loc) · 5.22 KB

CHANGELOG.md

File metadata and controls

206 lines (170 loc) · 5.22 KB

Change log

Unreleased

[1.10.6] - 2021-01-04

Added

  • added ndd to the anaconda conda-forge channel.

[1.10.3] - 2020-10-10

Fixed

  • fixed setuptools imports in setup.py

[1.10] - 2020-10-05

Changed

  • entropy function: removed the zk argument
  • the signature of the entropy function is: 
    entropy(nk, k=None, estimator='NSB', return_std=False)

[1.9.2] - 2020-09-21

Added

  • the entropy function handles mappings and generators as input counts.

[1.9.1] - 2020-07-07

Fixed

  • fix Counts.fit()

[1.9] - 2020-06-25

Changed

  • if argument k is omitted, the entropy function will guess a reasonable alphabet size and select the best estimator for the sampling regime.

[1.8.4] - 2020-06-24

Changed

  • the entropy function takes frequency counts (multiplicities) as input via the zk optional argument
  • renamed pk to nk
  • the signature of the entropy function is: 
    entropy(nk, k=None, zk=None, estimator='NSB', return_std=False)

[1.8.3] - 2020-06-16

Fixed

  • integration for large cardinalities

[1.8] - 2020-06-10

Added

  • full Bayesian error estimate (from direct computation of the posterior variance of the entropy)

[1.7] - 2020-06-09

Changed

  • rename version.json to package.json
  • check_pk: elements of input counts array are converted to int32 with no check
  • entropy() function: NSB estimator needs cardinality k as input
  • entropy() function: non-flat arrays are valid input counts

Fixed

  • robust integration range

[1.6.4] - 2020-05-07

Fixed

  • fix coinformation function for combinations of features
  • fix from_data for scalar inputs

[1.6.1] - 2019-08-11

Changed

For the NSB integration:

  • Find the saddle point maximizing p(alpha | data)
  • Set the integration range at 4 standard deviations around the saddle point
  • If the standard deviation is too small, return the entropy posterior mean at the saddle point

[1.6] - 2019-08-09

Added

  • MillerMadow estimator class
  • AsymptoticNSB estimator class
  • Grassberger estimator class

Changed

The signature of the entropy function has been changed to allow arbitrary entropy estimators. The new signature is

entropy(pk, k=None, estimator='NSB', return_std=False)

Check ndd.entropy_estimators for the available estimators.

[1.5] - 2019-08-02

Changed

For methods/functions working on data matrices: the default input is a n-by-p 2D array (n samples from p discrete variables, with different samples on different rows). Since release 1.3, the default was a transposed (p-by-n) data matrix. The behavior of functions taking frequency counts as input (e.g. the ndd.entropy function) is unchanged.

Added

[1.4] - 2019-05-20

Added

  • ndd.kullback_leibler_divergence

[1.3.2] - 2019-05-16

Changed

  • r (length of combinations) defaults to None

[1.3] - 2019-05-08

Changed

  • input data arrays must be p-by-n 2D ndarrays containing n samples from p discrete variables. This affects all methods/functions working directly on data:
    • histogram
    • from_data
    • interaction_information
    • coinformation
    • mutual_information
    • conditional_entropy

[1.2] - 2019-04-01

Changed

  • fixed conditional_entropy function
  • histogram: axis should be None is data matrix is tarnsposed.

[1.1] - 2019-03-26

Added

  • ndd.from_data
  • ndd.mutual_information
  • ndd.conditional_information
  • ndd.interaction_information
  • ndd.coinformation

[1.0] - 2019-03-19

Changed

  • Python3 only
  • fixed NumericError for valid entropy estimation

[0.9] - 2019-01-15

Added

  • jensen_shannnon_divergence() function.
  • estimators module.
  • Entropy, KLDivergence and JSDivergence estimators.

Changed

  • entropy() signature is now: entropy(pk, k=None, alpha=None, plugin=False, return_std=False)

[0.8] - 2019-01-08

Added

  • Entropy class.
  • from_data() returns an estimate from a data samples

Changed

  • histogram(): remove unique elements of array as optional output
  • histogram(): takes the axis and r optional args

[0.7.1] - 2018-06-22

Changed

  • entropy(): removed the (broken) function for estimator selection

[0.7] - 2018-03-22

Changed

  • entropy(): renamed argument 'dist' to 'plugin'

[0.6] - 2018-03-22

Changed

  • entropy(): renamed argument 'a' to 'alpha'

[0.5.5] - 2018-03-16

Changed

  • fix missing README.rst
  • handle input counts arrays > 1D (always flatten to 1D)
  • handle single-class case

[0.5.3] - 2018-03-15

Changed

  • require numpy >= 1.9

[0.5] - 2017-11-15

Changed

  • now a Python package
  • binary extension are installed in the package dir

[0.4.1] - 2017-11-14

Added

  • version.json

Removed

  • version.py

[0.4] - 2017-10-03

Changed

  • numerical integration is stable over large ranges of conditions (large k and/or large n)
  • raise an error if estimate is NaN
  • statndard implementation for ML estimator

Removed

  • preliminary notebooks

[0.3] - 2017-09-29

Added

  • Manifest.in

Changed

  • Changed return_error= to return_std= as key in the kwargs of ndd.entropy()
  • Renamed CHANGES.md to CHANGELOG.md
  • Check for NumPy version before setup